CN114781745A - Method, device and equipment for predicting mechanism deposit business condition - Google Patents
Method, device and equipment for predicting mechanism deposit business condition Download PDFInfo
- Publication number
- CN114781745A CN114781745A CN202210495857.XA CN202210495857A CN114781745A CN 114781745 A CN114781745 A CN 114781745A CN 202210495857 A CN202210495857 A CN 202210495857A CN 114781745 A CN114781745 A CN 114781745A
- Authority
- CN
- China
- Prior art keywords
- sequence data
- prediction
- data
- institution
- historical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007246 mechanism Effects 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000012937 correction Methods 0.000 claims abstract description 43
- 230000008520 organization Effects 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims description 87
- 238000012549 training Methods 0.000 claims description 42
- 238000012360 testing method Methods 0.000 claims description 20
- 238000010586 diagram Methods 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000001744 unit root test Methods 0.000 claims description 10
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 238000004590 computer program Methods 0.000 description 18
- 230000015654 memory Effects 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000012986 modification Methods 0.000 description 6
- 230000004048 modification Effects 0.000 description 6
- 238000012353 t test Methods 0.000 description 6
- 238000000692 Student's t-test Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000013475 authorization Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000001747 exhibiting effect Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 230000036961 partial effect Effects 0.000 description 2
- 101100134058 Caenorhabditis elegans nth-1 gene Proteins 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000002829 reductive effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Biology (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Computational Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Mathematical Optimization (AREA)
- Human Resources & Organizations (AREA)
- Mathematical Analysis (AREA)
- Technology Law (AREA)
- Probability & Statistics with Applications (AREA)
- Tourism & Hospitality (AREA)
- Algebra (AREA)
- Quality & Reliability (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure provides a method, a device and equipment for predicting the deposit business condition of an organization, which can be applied to the technical fields of artificial intelligence and finance. The method for predicting the deposit business condition of the institution comprises the following steps: acquiring mechanism deposit business condition time sequence data in a preset time period; differentiating the mechanism deposit business condition time sequence data to obtain stable sequence data and residual sequence data; inputting the stable sequence data into a first prediction sub-model in a mechanism deposit business condition prediction model, and outputting a first prediction result; inputting the residual sequence data into a second prediction submodel in the mechanism deposit business condition prediction model, and outputting a second prediction result, wherein the model structure of the second prediction submodel is different from that of the first prediction submodel; and inputting the first prediction result and the second prediction result into a result correction sub-model in the mechanism deposit business condition prediction model, and outputting third mechanism deposit business condition prediction information of a target time period.
Description
Technical Field
The present disclosure relates to the field of artificial intelligence and financial technologies, and in particular, to a method, an apparatus, a device, and a medium for predicting a deposit service status of an organization.
Background
With the increasingly competitive of the deposit business of the financial institutions, the change of the deposit business condition of the financial institutions is accurately predicted in time, and the institutions can conveniently adopt effective measures to prevent the loss of customers.
Due to the fact that factors influencing the deposit business state of the financial institution are more, for example: policy factors, market factors, service factors, product factors, client's own factors, and the like, the state of the financial institution's deposit transaction appears to change irregularly. In the related art, generally, a certain historical time point is used for prediction based on linear characteristics of deposit business data, so that the accuracy of a prediction result is low.
Disclosure of Invention
In view of the above problems, the present disclosure provides a method, apparatus, device, and medium for predicting the status of an institution deposit transaction.
According to one aspect of the disclosure, a method for predicting institution deposit business status is provided, comprising:
acquiring mechanism deposit business condition time sequence data in a preset time period;
differentiating the mechanism deposit business condition time sequence data to obtain stable sequence data and residual sequence data;
inputting the stable sequence data into a first prediction sub-model in the mechanism deposit business condition prediction model, and outputting a first prediction result, wherein the first prediction result represents the first mechanism deposit business condition prediction information of a target time period;
inputting the residual sequence data into a second prediction sub-model in the mechanism deposit business condition prediction model, and outputting a second prediction result, wherein the second prediction result represents second mechanism deposit business condition prediction information of a target time period, and the model structure of the second prediction sub-model is different from that of the first prediction sub-model; and
and inputting the first prediction result and the second prediction result into a result correction sub-model in the mechanism deposit business condition prediction model, and outputting third mechanism deposit business condition prediction information of a target time period.
According to the embodiment of the disclosure, the mechanism deposit business state time sequence data is subjected to difference processing to obtain stable sequence data and residual sequence data, and the method comprises the following steps:
performing differential processing on the mechanism deposit business condition time sequence data to obtain target sequence data, wherein the target sequence data comprises sequence data with stable fluctuation;
calculating target test statistic of the target sequence data by using a unit root test algorithm;
and classifying the time sequence data of the deposit business condition according to the target test statistic to obtain stable sequence data and residual sequence data.
According to the embodiment of the disclosure, differential processing is performed on the mechanism deposit business condition time sequence data to obtain target sequence data, and the differential processing comprises the following steps:
carrying out differential processing on the mechanism deposit business condition time sequence data to obtain a target time sequence diagram;
and determining the time sequence data which presents stable fluctuation in the target time sequence chart as the target sequence data.
According to the embodiment of the disclosure, the training method of the mechanism deposit business condition prediction model comprises the following steps:
acquiring a sample data set, wherein the sample data set comprises sample data of a plurality of historical preset time periods, and the sample data comprises time sequence data of historical institution deposit service conditions and actual institution deposit service condition data of historical target time periods;
for each sample data, performing differential processing on the time sequence data of the historical deposit business condition to obtain historical stable sequence data and historical residual sequence data;
training a first initial sub-model according to the historical steady sequence data and the actual deposit business condition data of the organization to obtain a trained first prediction sub-model;
training a second initial sub-model according to the historical residual sequence data and the actual deposit business condition data of the organization to obtain a trained second prediction sub-model;
inputting the historical steady sequence data into a first prediction sub-model, and outputting first historical institution deposit service condition prediction data;
inputting the historical residual sequence data into a second prediction submodel, and outputting second historical institution deposit service condition prediction data;
and training a third initial sub-model according to the first historical institution deposit business condition prediction data, the second historical institution deposit business condition prediction data and the institution actual deposit business condition data to obtain a trained result correction sub-model.
According to the embodiment of the disclosure, training a third initial sub-model according to the first historical institution deposit business condition prediction data, the second historical institution deposit business condition prediction data and the institution actual deposit business condition data to obtain a trained result correction sub-model, and the training comprises the following steps:
inputting the predicted data of the deposit business condition of the first historical institution and the predicted data of the deposit business condition of the second historical institution into a third initial submodel, and outputting the predicted data of the deposit business condition of the third historical institution;
according to the third history institution deposit business condition prediction data and the institution actual deposit business condition data, determining a prediction result error;
and adjusting the model parameters of the third initial submodel according to the error of the predicted result to obtain a result correction submodel.
According to the embodiment of the disclosure, the method for obtaining the result correction submodel by adjusting the model parameters of the third initial submodel according to the error of the prediction result comprises the following steps:
determining the model parameter of the third initial sub-model as a target model parameter under the condition that the error of the prediction result is smaller than a preset threshold value;
and determining a result correction submodel according to the target model parameters.
According to the embodiment of the disclosure, the differential processing is performed on the historical deposit business condition time sequence data to obtain historical stationary sequence data and historical residual sequence data, and the differential processing comprises the following steps:
differential processing is carried out on the time sequence data of the deposit service condition of the historical organization to obtain target historical sequence data, wherein the target historical sequence data comprise sequence data with stable fluctuation;
calculating target test statistics of the target historical sequence data by using a unit root test algorithm;
and classifying the time sequence data of the deposit service condition of the historical organization according to the target test statistic to obtain historical stable sequence data and historical residual sequence data.
Another aspect of the present disclosure provides an apparatus for predicting an institution deposit transaction condition, comprising: the device comprises an acquisition module, a processing module, a first prediction module, a second prediction module and a correction module. The acquisition module is used for acquiring the mechanism deposit business condition time sequence data in the preset time period. And the processing module is used for carrying out differential processing on the mechanism deposit business condition time sequence data to obtain stable sequence data and residual sequence data. The first prediction module is used for inputting the stable sequence data into a first prediction sub-model in the mechanism deposit business condition prediction model and outputting a first prediction result, wherein the first prediction result represents the first mechanism deposit business condition prediction information of a target time period; the second prediction module is used for inputting the residual sequence data into a second prediction sub-model in the mechanism deposit business condition prediction model and outputting a second prediction result, wherein the second prediction result represents second mechanism deposit business condition prediction information of a target time period, and the model structure of the second prediction sub-model is different from that of the first prediction sub-model; and the correction module is used for inputting the first prediction result and the second prediction result into a result correction sub-model in the mechanism deposit business condition prediction model and outputting third mechanism deposit business condition prediction information of a target time period.
According to an embodiment of the present disclosure, the processing module includes a first processing sub-module, a calculation sub-module, and a classification sub-module. The first processing submodule is used for carrying out differential processing on the mechanism deposit business condition time sequence data to obtain target sequence data, wherein the target sequence data comprise sequence data with stable fluctuation. And the calculation submodule is used for calculating the target test statistic of the target sequence data by using a unit root test algorithm. And the classification submodule is used for classifying the deposit business condition time sequence data according to the target test statistic to obtain stable sequence data and residual sequence data.
According to an embodiment of the present disclosure, the first processing submodule includes a first processing unit and a first determination unit. The first processing unit is used for carrying out differential processing on the mechanism deposit business state time sequence data to obtain a target time sequence diagram. A first determination unit configured to determine, as the target sequence data, sequence data exhibiting a stable fluctuation in the target sequence diagram.
According to an embodiment of the present disclosure, the training module includes an acquisition sub-module, a second processing sub-module, a first training sub-module, a second training sub-module, a first prediction sub-module, a second prediction sub-module, and a third training sub-module. The acquisition submodule is used for acquiring a sample data set, wherein the sample data set comprises sample data of a plurality of historical preset time periods, and the sample data comprises time sequence data of the deposit service condition of a historical institution and actual deposit service condition data of the institution in a historical target time period. And the second processing submodule is used for carrying out differential processing on the time sequence data of the historical deposit service condition aiming at each sample data to obtain historical stable sequence data and historical residual error sequence data. And the first training submodule is used for training a first initial submodel according to the historical stable sequence data and the actual deposit business condition data of the organization to obtain a trained first prediction submodel. And the second training submodule is used for training a second initial submodel according to the historical residual sequence data and the actual deposit business condition data of the organization to obtain a trained second prediction submodel. And the first prediction sub-module is used for inputting the historical stable sequence data into a first prediction sub-model and outputting the prediction data of the deposit business condition of the first historical institution. And the second prediction sub-module is used for inputting the historical residual sequence data into a second prediction sub-model and outputting second historical institution deposit service condition prediction data. And the third training submodule is used for training a third initial submodel according to the first historical institution deposit business condition prediction data, the second historical institution deposit business condition prediction data and the institution actual deposit business condition data to obtain a trained result correction submodel.
According to an embodiment of the present disclosure, the second processing submodule includes a second processing unit, a calculation unit, and a classification unit. The second processing unit is used for carrying out difference processing on the historical institution deposit business condition time sequence data to obtain target historical sequence data, wherein the target historical sequence data comprises sequence data with stable fluctuation. And the calculating unit is used for calculating the target test statistic of the target historical sequence data by using a unit root test algorithm. And the classification unit is used for classifying the time sequence data of the deposit service condition of the historical institution according to the target test statistic to obtain historical stationary sequence data and historical residual sequence data.
According to an embodiment of the present disclosure, the third training submodule includes a first prediction unit, a second determination unit, and an adjustment unit. And the first prediction unit is used for inputting the first historical institution deposit business condition prediction data and the second historical institution deposit business condition prediction data into a third initial sub-model and outputting third historical institution deposit business condition prediction data. And the second determining unit is used for determining a prediction result error according to the third history institution deposit business condition prediction data and institution actual deposit business condition data. And the adjusting unit is used for adjusting the model parameters of the third initial submodel according to the prediction result error to obtain a result correction submodel.
According to an embodiment of the present disclosure, the adjustment unit includes a first determination subunit and a second determination subunit. The first determining subunit is configured to determine, as the target model parameter, the model parameter of the third initial sub-model when the prediction result error is smaller than a preset threshold. And the second determining subunit is used for determining a result and correcting the submodel according to the target model parameters.
Another aspect of the present disclosure provides an electronic device including: one or more processors; a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the above method for predicting the institution credit business condition.
Another aspect of the present disclosure also provides a computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the above-described method for predicting a status of a mechanism deposit transaction.
Another aspect of the present disclosure also provides a computer program product comprising a computer program which, when executed by a processor, implements the above method for predicting the status of the institution deposit transaction.
According to the embodiment of the disclosure, steady sequence data and residual sequence data are obtained by carrying out differential processing on mechanism deposit business state time sequence data in a preset time period, the steady sequence data and the residual sequence data are respectively input into two prediction submodels with different structures, two mechanism deposit business state prediction results in a target time period are obtained, and then the two mechanism deposit business state prediction results are adjusted through a result correction submodel, so that final mechanism deposit business state prediction information in the target time period is obtained. After differential processing is carried out on the mechanism deposit business condition time sequence data, the randomly fluctuant mechanism deposit business condition time sequence data are divided into stable sequence data and residual sequence data, and a prediction result obtained by using the stable sequence data and a prediction result obtained by using the residual sequence data are corrected, so that the randomness of the residual sequence data is weakened, and the accuracy of the mechanism deposit business condition prediction result is improved.
Drawings
The foregoing and other objects, features and advantages of the disclosure will be apparent from the following description of embodiments of the disclosure, which proceeds with reference to the accompanying drawings, in which:
FIG. 1 schematically illustrates an application example architecture diagram of a method, apparatus, device, medium and program product for forecasting institution credit business conditions in accordance with embodiments of the present disclosure;
FIG. 2 schematically illustrates a flow chart of a method for predicting institution deposit business conditions in accordance with an embodiment of the present disclosure;
FIG. 3 schematically illustrates a flow chart of a method of processing institution credit business condition timing data in accordance with an embodiment of the disclosure;
FIG. 4 schematically illustrates a flow chart of a training method of an institution deposit business condition prediction model according to an embodiment of the disclosure;
FIG. 5 is a block diagram schematically illustrating the structure of an apparatus for predicting the status of the institution deposit service in accordance with an embodiment of the present disclosure; and
FIG. 6 schematically illustrates a block diagram of an electronic device suitable for implementing a method for predicting institution credit business conditions in accordance with an embodiment of the present disclosure.
Detailed Description
Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is illustrative only and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.
Where a convention analogous to "at least one of A, B and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B and C" would include but not be limited to systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).
It should be noted that the method and the device for predicting the institution deposit business condition disclosed by the invention can be used in the technical fields of artificial intelligence and finance, and can also be used in any fields except the finance field.
In the technical scheme of the disclosure, the collection, storage, use, processing, transmission, provision, disclosure, application and other processing of the personal information of the related user are all in accordance with the regulations of related laws and regulations, necessary confidentiality measures are taken, and the customs of the public order is not violated.
In the technical scheme of the disclosure, before the personal information of the user is acquired or collected, the authorization or the consent of the user is acquired.
The embodiment of the disclosure provides a method for predicting mechanism deposit business conditions, which includes the steps of carrying out differential processing on mechanism deposit business condition time sequence data in a preset time period to obtain stable sequence data and residual sequence data, inputting the stable sequence data and the residual sequence data into two prediction sub models with different structures respectively to obtain prediction results of the mechanism deposit business conditions in a target time period, and adjusting the prediction results of the mechanism deposit business conditions through a result correction sub model to obtain final mechanism deposit business condition prediction information in the target time period. After differential processing is carried out on the mechanism deposit business condition time sequence data, the mechanism deposit business condition time sequence data which fluctuates randomly are divided into stable sequence data and residual sequence data, and a prediction result obtained by the stable sequence data and a prediction result obtained by the residual sequence data are used for correction, so that the randomness of the residual sequence data is weakened, and the accuracy of the mechanism deposit business condition prediction result is improved.
Fig. 1 schematically shows an application example architecture diagram of a method for predicting the status of the institution deposit business according to an embodiment of the present disclosure.
As shown in fig. 1, an exemplary architecture 100 of an application according to this embodiment may include terminal devices 101, 102, 103, a network 104, a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.
A user may use terminal devices 101, 102, 103 to interact with a server 105 over a network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have installed thereon various communication client applications, such as shopping-like applications, web browser applications, search-like applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only).
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 105 may be a server providing various services, such as a background management server (for example only) providing support for websites browsed by users using the terminal devices 101, 102, 103. The backend management server may analyze and process the received data such as the user request, and feed back a processing result (for example, a web page, information, or data obtained or generated according to the user request) to the terminal device.
It should be noted that the method for predicting the deposit service status of the institution provided by the embodiment of the present disclosure may be generally executed by the server 105. Accordingly, the device for predicting the institution deposit business condition provided by the embodiment of the present disclosure may be generally disposed in the server 105. The method for predicting the institution deposit service condition provided by the embodiment of the disclosure may also be executed by a server or a server cluster which is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and/or the server 105. Accordingly, the device for predicting the institution deposit business condition provided by the embodiment of the present disclosure may also be disposed in a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103 and/or the server 105.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for an implementation.
The method for predicting the institution deposit business condition of the disclosed embodiment will be described in detail below through fig. 2 to 4 based on the scenario described in fig. 1.
FIG. 2 schematically shows a flow chart of a method for predicting the status of an institution's deposit transaction in accordance with an embodiment of the disclosure.
As shown in fig. 2, the method for predicting the condition of the institution deposit transaction of this embodiment includes operations S210 to S250.
In operation S210, institution deposit transaction status time series data within a preset time period is acquired.
In embodiments of the present disclosure, consent or authorization of the user may be obtained prior to obtaining the user's information. For example, a request for obtaining user information may be issued to the user before operation S220. In case that the user information can be acquired with the user' S consent or authorization, the operation S220 is performed.
According to an embodiment of the present disclosure, the preset time period may be any time period before the target time period, for example: the target time period may be the nth year, the preset time period may be the nth-1 year, or the nth-2 year, or the nth-3 year, etc. The target period may be the mth month, and the preset period may be the m-1 st year, or the m-2 nd year, or the m-3 rd year, etc. The target time period may also be measured in days.
According to embodiments of the present disclosure, institution credit business condition time series data may include sequence data for changes in an institution credit balance over time.
In operation S220, the mechanism deposit service status time series data is subjected to difference processing to obtain stable sequence data and residual sequence data.
According to the embodiment of the disclosure, the difference processing is performed on the institution deposit business condition time series data, which may include first order difference processing, second order difference processing and the like, until a partial sequence in the differentiated institution deposit business condition time series data shows a stable fluctuation phenomenon, which indicates that stable sequence data and residual sequence data exist in the institution deposit business condition time series data, and the stable sequence data and the residual sequence data are obtained.
In operation S230, the stationary sequence data is input into a first prediction submodel in the institution deposit business condition prediction model, and a first prediction result is output, where the first prediction result represents first institution deposit business condition prediction information of the target time period.
According to embodiments of the present disclosure, the first predictor model may include a differential Integrated Moving Average regression ARIMA model (autoregestive Integrated Moving Average model). The ARIMA model comprises three model parameters, namely an autoregressive term number p, a moving average maturity number q and a difference number d which is used for enabling time sequence data of the institution deposit service condition to become stable sequence data.
According to an embodiment of the present disclosure, for example: the initial model parameters in the differential integrated moving average sub-regression ARIMA model may be determined according to a minimum information criterion (AIC criterion), such as: and when the function is minimum according to the minimum information criterion, determining initial model parameters p, d and q in the difference integration moving average sub-regression ARIMA model. Since the modeling process of the difference integration moving average sub-regression ARIMA model is a mature technology, it is not described herein any further.
According to the embodiment of the disclosure, the stationary sequence data is input into a differential integrated moving average autoregressive (ARIMA) model, and a first prediction result is output. For example: the first prediction result may be a first institution deposit balance value for the target time period.
In operation S240, the residual sequence data is input into a second predictor model in the institution deposit business condition prediction model, and a second prediction result is output, wherein the second prediction result represents second institution deposit business condition prediction information of the target time period, and the second predictor model has a different model structure from the first predictor model.
According to an embodiment of the present disclosure, the second predictor model may comprise a grey model GM (1, 1). Wherein the first 1 in brackets represents a first order differential equation and the second 1 represents a differential equation having one variable.
According to the embodiment of the disclosure, due to factors influencing the random change of the deposit business condition of the institution, for example: policy factors, market factors, service factors, product factors, client factors and the like all have irregular changes, and a gray model can be constructed according to the factors.
According to the embodiment of the present disclosure, the residual sequence data X obtained by the difference may be(0)Is set as shown in formula (1):
X(0)={x(0)(1),x(0)(2),...,x(0)(n)} (1)
according to the embodiment of the present disclosure, the residual sequence data shown in equation (1) may be accumulated to obtain the first-order accumulation generation sequence X shown in equation (2)(1):
X(1)={x(1)(1),x(1)(2),...,x(1)(n)} (2)
according to an embodiment of the present disclosure, a difference equation as shown in equation (4) and an ordinary differential equation as shown in equation (5) of the GM (1, 1) model are constructed.
x(0)(k)+az(1)(k)=b,(k=2,3...,n) (4)
Where a in equations (4) and (5) represents a coefficient of development, which is determined by policy factors, market factors, service factors, product factors, customer's own factors, etc., which affect the status of the institution deposit transaction in the embodiment of the present disclosure. In the formulae (4) and (5), b represents the amount of ash action. Z in the formula (4)(1)(k) Generating a sequence X for first order accumulation(1)Is represented by equation (6):
z(1)(k)=0.5x(1)(k)+0.5x(1)(k-1)k=2,3,...,n (7)
according to an embodiment of the present disclosure, solving equation (5) may result:
according to the embodiment of the disclosure, the serial data obtained by equation (8) is subtracted and reduced to obtain the forecast result x of the deposit business condition of the second institution(0)(k +1) as shown in formula (9):
x(0)(k+1)=x(1)(k+1)-x(1)(k)k=2,3,...,n (9)
according to the embodiment of the present disclosure, the GM (1, 1) model may be established according to the above procedure, and the modeling accuracy of the GM (1, 1) model may be calculated by a residual value verification method. And determining that the precision of the established GM (1, 1) model is qualified under the condition that the error probability of the result obtained by calculation through a residual value test method is greater than 95%.
According to an embodiment of the present disclosure, the residual sequence data is input into the gray model GM (1, 1), and the second prediction result is output. For example: the second prediction result may be a second institution deposit balance value for the target time period.
In operation S250, the first prediction result and the second prediction result are input to a result correction sub-model in the institution deposit business condition prediction model, and third institution deposit business condition prediction information of the target time period is output.
According to an embodiment of the present disclosure, in the result modification sub-model, the third institution deposit transaction condition prediction information Y of the modified final target time period may be as shown in equation (10):
Y=mA+B (10)
wherein Y represents the corrected third institution deposit service condition prediction information of the final target time period, A represents the first prediction result, B represents the second prediction result, and m represents the correction correlation coefficient.
Wherein A ═ a1,a2,...aj} (11)
Wherein, ajAnd a first prediction result showing the institution deposit business condition of the nth year obtained by prediction by using the stable sequence data in the institution deposit business condition time sequence data of the jth year.
Wherein B ═ B1,b2,...bj} (12)
Wherein, bjAnd a second prediction result showing the institution deposit business condition of the nth year obtained by prediction by using residual sequence data in the institution deposit business condition time sequence data of the jth year.
According to an embodiment of the present disclosure, for a trained result-modifying submodel, modifying the correlation coefficients is determined. Therefore, the first prediction result and the second prediction result are input into the result correction sub-model, and the third institution deposit business condition prediction information of the target time period can be obtained.
According to the embodiment of the disclosure, steady sequence data and residual sequence data are obtained by carrying out differential processing on mechanism deposit business condition time sequence data in a preset time period, the steady sequence data and the residual sequence data are respectively input into two prediction sub models with different structures, so that two mechanism deposit business condition prediction results in a target time period are obtained, and then the two mechanism deposit business condition prediction results are adjusted through a result correction sub model, so that final mechanism deposit business condition prediction information in the target time period is obtained. After differential processing is carried out on the mechanism deposit business condition time sequence data, the randomly fluctuant mechanism deposit business condition time sequence data are divided into stable sequence data and residual sequence data, and a prediction result obtained by using the stable sequence data and a prediction result obtained by using the residual sequence data are corrected, so that the randomness of the residual sequence data is weakened, and the accuracy of the mechanism deposit business condition prediction result is improved.
FIG. 3 schematically shows a flowchart of a method for processing institution credit transaction status timing data in accordance with an embodiment of the disclosure.
As shown in FIG. 3, the method of processing institution deposit transaction status time series data of this embodiment includes operations S310 to S330.
In operation S310, the institution deposit transaction status time-series data is subjected to difference processing to obtain target sequence data, where the target sequence data includes sequence data with stable fluctuation.
According to an embodiment of the present disclosure, the differential processing may include first order difference, second order difference, third order difference, and the like until occurrence of sequence data in which local fluctuation is stable.
In operation S320, a target test statistic of the target sequence data is calculated using a unit root test algorithm.
According to an embodiment of the present disclosure, for example: t-test statistics for the target sequence data can be calculated using a t-test method.
In operation S330, the mechanism deposit service status time series data is classified according to the target test statistic, and stable sequence data and residual sequence data are obtained.
According to an embodiment of the present disclosure, for example: the result reliability in statistics can be used as a judgment index, and when the result reliability of the t-test statistic of the calculated target sequence data is greater than or close to a preset threshold, the target sequence data is represented as residual sequence data. And when the result reliability of the t test statistic of the calculated target sequence data is obviously smaller than a preset threshold value, representing that the target sequence data is stable sequence data.
According to the embodiment of the disclosure, the unit root inspection algorithm is utilized to divide the mechanism deposit business state time sequence data into the stable sequence data and the residual sequence data, the time sequence data are processed in the early stage of model prediction, and then the models with different structures are utilized to predict the sequence data with different types, so that the problem of low result accuracy caused by prediction only by adopting the sequence data with a single type in the related technology is solved.
According to the embodiment of the disclosure, differential processing is performed on the mechanism deposit business condition time sequence data to obtain target sequence data, and the differential processing comprises the following steps:
carrying out differential processing on the mechanism deposit business condition time sequence data to obtain a target time sequence diagram;
and determining the time sequence data which presents stable fluctuation in the target time sequence chart as the target sequence data.
According to an embodiment of the present disclosure, for example: and performing first-order difference and second-order difference on the mechanism deposit business condition time sequence data until a target time sequence chart with part of stable random fluctuation phenomena is obtained. And intercepting time sequence data presenting stable fluctuation in the target time sequence chart, and determining the target sequence data for unit root inspection to determine the target sequence data for classification.
According to the embodiment of the disclosure, steady time sequence data and residual error time sequence data used for predicting the mechanism deposit business state can be extracted from random and irregular mechanism deposit business state time sequence data by continuously carrying out differential processing on the mechanism deposit business state time sequence data until partial stably fluctuating time sequence data are obtained and then classifying the time sequence data, so that the data processing efficiency is improved.
FIG. 4 schematically shows a flowchart of a training method of an institution deposit business condition prediction model according to an embodiment of the disclosure.
As shown in fig. 4, the training method of the institution deposit business condition prediction model of this embodiment includes operations S410 to S470.
In operation S410, a sample data set is obtained, where the sample data set includes sample data of a plurality of historical preset time periods, where the sample data includes time series data of historical institution deposit service conditions and actual institution deposit service condition data of historical target time periods.
According to an embodiment of the disclosure, the historical institution deposit service condition time series data may include the deposit balance time series data of institution a of year n-1, year n-2, year n-3.
In operation S420, for each sample data, the historical deposit service condition time series data is subjected to difference processing, so as to obtain historical stationary sequence data and historical residual sequence data.
According to an embodiment of the present disclosure, for example: the deposit balance time sequence data of the institution A of the (n-1) th year is subjected to difference processing, so that stable sequence data of the deposit balance of the institution A of the (n-1) th year and residual sequence data of the deposit balance of the institution A of the (n-1) th year can be obtained.
In operation S430, a first initial sub-model is trained according to the historical stationary sequence data and the actual deposit business condition data of the organization, so as to obtain a trained first prediction sub-model.
According to an embodiment of the present disclosure, for example: the stable sequence data of the deposit balance of the institution A of the (n-1) th year can be input into a first initial sub-model, and the predicted value M of the deposit balance of the institution A of the (n) th year is output1And calculating a predicted deposit balance value M of the institution A of the nth year1And if the relative error between the actual deposit balance value of the institution A in the nth year and the actual deposit balance value is larger than 1%, adjusting the model parameters of the first initial sub-model. For example: model parameters p, d, q in ARIMA (p, d, q) model until the relative error is less than 1%. Inputting the steady sequence data of the deposit balance of the institution A of the (n-2) th year into a first initial sub-model after adjusting the model parameters, and outputting a predicted value M of the deposit balance of the institution A of the (n) th year2And calculating a predicted value M of the deposit balance of the nth organization A2Relative error with the actual value of the deposit balance of the institution A in the nth year, and repeating the steps until the stable sequence data of the deposit balance of the institution A in any year is input into the first prediction sub-model, and the predicted value M of the deposit balance of the institution A in the nth year is outputnAnd finishing the training if the relative errors with the actual value of the deposit balance of the institution A in the nth year are less than 1 percent.
In operation S440, a second initial sub-model is trained according to the historical residual sequence data and the actual deposit business condition data of the organization, so as to obtain a trained second prediction sub-model.
According to the embodiment of the disclosure, residual sequence data of the deposit balance of the institution a of the (n-1) th year can be input into the second initial sub-model, and the deposit balance predicted value Q of the institution a of the (n) th year is output1And calculating a predicted deposit balance value Q of the institution A of the nth year1Balance of deposit with the nth organization AAnd if the error probability of the actual value is less than 60%, adjusting the model parameters of the second initial sub-model. For example: model parameters a and b in GM (1, 1) model until the error probability is greater than 95%. Inputting residual sequence data of the deposit balance of the institution A of the nth-2 years into a second initial sub-model after adjusting model parameters, and outputting a predicted value Q of the deposit balance of the institution A of the nth year2And calculating a predicted value Q of the deposit balance of the institution A of the nth year2The error probability with the actual value of the deposit balance of the institution A in the nth year is circulated until residual sequence data of the deposit balance of the institution A in any year is input into a first prediction sub-model, and a predicted value Q of the deposit balance of the institution A in the nth year is outputnAnd if the error probability of the actual deposit balance value of the institution A in the nth year is more than 95 percent, finishing the training.
In operation S450, the historical stationary sequence data is input into a first prediction submodel, and first historical institution deposit transaction condition prediction data is output.
According to an embodiment of the present disclosure, for example: inputting the steady sequence data of the institution A in the (n-1) th year, the (n-2) th year and the (n-3) th year into a first prediction sub-model, and outputting the prediction data of the deposit business condition of the first historical institution. The first historical institution deposit service condition prediction data may be a deposit balance prediction value sequence of the institution a of the nth year, for example: (a)1,a2,a3,a4...ai)。
In operation S460, the historical residual sequence data is input to a second prediction submodel, and second historical agency deposit service condition prediction data is output.
According to an embodiment of the present disclosure, for example: for example: inputting residual sequence data of the institution A in the n-1 th year, the n-2 th year and the n-3 th year. The second historical institution deposit service condition prediction data may be a deposit balance prediction value sequence of the institution a of the nth year, for example: (b)1,b2,b3,b4...bi)。
In operation S470, the third initial sub-model is trained according to the first historical institution deposit business status prediction data, the second historical institution deposit business status prediction data, and the institution actual deposit business status data, to obtain a trained result correction sub-model.
According to an embodiment of the present disclosure, for example: deposit business condition prediction data (a) of a first historical institution1,a2,a3,a4...ai) Second historical institution deposit business condition prediction data (b)1,b2,b3,b4...bi) And actual deposit business state data p of organization0And training the third initial submodel to obtain a trained result correction submodel.
According to the embodiment of the disclosure, the actual institution deposit business condition data and the institution deposit business condition data in the historical preset time period are respectively trained to obtain the first prediction sub-model and the second prediction sub-model, the sub-models are corrected by utilizing a plurality of prediction result training results output by the trained first prediction sub-model and the trained second prediction sub-model, the institution deposit business condition can be predicted by utilizing residual sequence data, the randomness of the residual sequence data is weakened, and the model training efficiency is improved.
According to the embodiment of the disclosure, a third initial submodel is trained according to the first historical institution deposit business condition prediction data, the second historical institution deposit business condition prediction data and the institution actual deposit business condition data, and a trained result correction submodel is obtained, wherein the method comprises the following steps:
inputting the predicted data of the deposit business condition of the first historical institution and the predicted data of the deposit business condition of the second historical institution into a third initial submodel, and outputting the predicted data of the deposit business condition of the third historical institution;
according to the third history institution deposit business condition prediction data and the institution actual deposit business condition data, determining a prediction result error;
and adjusting the model parameters of the third initial submodel according to the error of the predicted result to obtain a result correction submodel.
According to embodiments of the present disclosureSuch as: the first deposit balance predicted value a of the institution A of the year n, which is predicted from the stationary sequence data in the institution deposit balance time series data of the institution A of the year n-1, can be obtained1And a second deposit balance predicted value b of the institution A of the nth year, which is obtained by prediction according to residual sequence data in institution deposit balance time sequence data of the institution A of the (n-1) th year1Inputting the third deposit balance predicted value q of the nth mechanism A into the third initial submodel1。
According to an embodiment of the present disclosure, for example: can be predicted by calculating the third deposit balance q of the nth-year organization A1Actual value p of deposit balance with the nth organization A0And determining a prediction result error.
According to the embodiment of the disclosure, the method for obtaining the result correction submodel by adjusting the model parameters of the third initial submodel according to the error of the prediction result comprises the following steps:
determining the model parameter of the third initial sub-model as a target model parameter under the condition that the error of the prediction result is smaller than a preset threshold value;
and determining a result correction submodel according to the target model parameters.
According to an embodiment of the present disclosure, for example: when the error of the prediction result is greater than 5%, the model parameter of the third initial sub-model needs to be adjusted, and the model parameter of the third initial sub-model may be a modified correlation coefficient. When the result (a) is predicted for any one groupi,bi) Inputting a third deposit balance predicted value q of the nth institution A obtained by a third initial submodeliActual value p of deposit balance with the nth organization A0When the predicted result errors are less than 5%, it can be determined that the result correction submodel has been trained, and at this time, the correction correlation coefficient of the result correction submodel is the target model parameter.
According to the embodiment of the disclosure, the correction correlation coefficients of the first prediction result, the second prediction result and the actual data of the institution deposit business condition are determined through the first prediction result of a large amount of stable sequence data, the second prediction result of a residual sequence and the actual data of the institution deposit business condition, the training of the result correction submodel is completed, and under the condition that the model precision of the two prediction submodels is high, the correlation between the two prediction results and the actual data can be corrected by using the actual data of the institution deposit business condition together, so that the prediction accuracy of the result correction submodel is improved.
According to the embodiment of the disclosure, the differential processing is performed on the historical deposit business condition time sequence data to obtain historical stationary sequence data and historical residual sequence data, and the differential processing comprises the following steps:
differential processing is carried out on the time sequence data of the deposit service condition of the historical organization to obtain target historical sequence data, wherein the target historical sequence data comprise sequence data with stable fluctuation;
calculating target test statistics of the target historical sequence data by using a unit root test algorithm;
and classifying the time sequence data of the deposit service condition of the historical institution according to the target test statistic to obtain historical stationary sequence data and historical residual sequence data.
According to an embodiment of the present disclosure, for example: the result reliability in statistics can be used as a judgment index, and when the result reliability of the t-test statistic of the calculated target historical sequence data is greater than or close to a preset threshold value, the target historical sequence data is represented as historical residual sequence data. And when the result reliability of the t-test statistic of the calculated target historical sequence data is obviously smaller than a preset threshold, representing that the target historical sequence data is historical stable sequence data.
According to the embodiment of the disclosure, the unit root inspection algorithm is utilized to divide the time sequence data of the deposit business condition of the historical organization into the historical stable sequence data and the historical residual sequence data, different types of sequence data are utilized to train models with different structures, and the problem of low result accuracy caused by prediction only by adopting single type of sequence data in the related technology is solved.
Based on the method for predicting the institution deposit business condition, the disclosure also provides a device for predicting the institution deposit business condition. The apparatus will be described in detail below with reference to fig. 5.
Fig. 5 is a block diagram schematically illustrating the structure of an apparatus for predicting the status of the institution deposit service according to an embodiment of the present disclosure.
As shown in fig. 5, the apparatus 500 for predicting the condition of the institution deposit service of this embodiment includes: an acquisition module 510, a processing module 520, a first prediction module 530, a second prediction module 540, and a correction module 550.
The obtaining module 510 is configured to obtain mechanism deposit business status time sequence data within a preset time period. In an embodiment, the obtaining module 510 may be configured to perform the operation S210 described above, which is not described herein again.
And the processing module 520 is configured to perform difference processing on the mechanism deposit service status time sequence data to obtain stable sequence data and residual sequence data. In an embodiment, the processing module 520 may be configured to perform the operation S220 described above, which is not described herein again.
The first prediction module 530 is configured to input the stationary sequence data into a first prediction sub-model in the mechanism deposit business condition prediction model, and output a first prediction result, where the first prediction result represents first mechanism deposit business condition prediction information of the target time period. In an embodiment, the first prediction module 530 may be configured to perform the operation S230 described above, which is not described herein again.
And a second prediction module 540, configured to input the residual sequence data into a second prediction sub-model in the mechanism deposit business condition prediction model, and output a second prediction result, where the second prediction result represents second mechanism deposit business condition prediction information of the target time period, and a model structure of the second prediction sub-model is different from that of the first prediction sub-model. In an embodiment, the second prediction module 540 may be configured to perform the operation S240 described above, and is not described herein again.
And a correcting module 550, configured to input the first prediction result and the second prediction result into a result correction sub-model in the institution deposit business condition prediction model, and output third institution deposit business condition prediction information in the target time period. In an embodiment, the modification module 550 may be configured to perform the operation S250 described above, and is not described herein again.
According to an embodiment of the present disclosure, the processing module includes a first processing sub-module, a calculation sub-module, and a classification sub-module. The first processing submodule is used for carrying out differential processing on the mechanism deposit business condition time sequence data to obtain target sequence data, wherein the target sequence data comprise sequence data with stable fluctuation. And the calculation submodule is used for calculating the target test statistic of the target sequence data by using a unit root test algorithm. And the classification submodule is used for classifying the deposit business condition time sequence data according to the target test statistic to obtain stable sequence data and residual sequence data.
According to an embodiment of the present disclosure, the first processing submodule includes a first processing unit and a first determination unit. The first processing unit is used for carrying out differential processing on the mechanism deposit business state time sequence data to obtain a target time sequence diagram. A first determination unit configured to determine, as the target sequence data, sequence data exhibiting a stable fluctuation in the target sequence diagram.
According to an embodiment of the present disclosure, the training module includes an acquisition sub-module, a second processing sub-module, a first training sub-module, a second training sub-module, a first prediction sub-module, a second prediction sub-module, and a third training sub-module. The acquisition submodule is used for acquiring a sample data set, wherein the sample data set comprises sample data of a plurality of historical preset time periods, and the sample data comprises time sequence data of historical institution deposit service conditions and actual institution deposit service condition data of historical target time periods. And the second processing submodule is used for carrying out differential processing on the time sequence data of the historical deposit service condition aiming at each sample data to obtain historical stable sequence data and historical residual sequence data. And the first training submodule is used for training a first initial submodel according to the historical stable sequence data and the actual deposit business condition data of the organization to obtain a trained first prediction submodel. And the second training submodule is used for training a second initial submodel according to the historical residual sequence data and the actual deposit business condition data of the organization to obtain a trained second prediction submodel. And the first prediction sub-module is used for inputting the historical stable sequence data into the first prediction sub-model and outputting the first historical institution deposit business condition prediction data. And the second prediction submodule is used for inputting the historical residual sequence data into a second prediction submodel and outputting second historical institution deposit business condition prediction data. And the third training submodule is used for training a third initial submodel according to the first historical institution deposit business condition prediction data, the second historical institution deposit business condition prediction data and the institution actual deposit business condition data to obtain a trained result correction submodel.
According to an embodiment of the present disclosure, the second processing submodule includes a second processing unit, a calculation unit, and a classification unit. The second processing unit is used for carrying out difference processing on the historical mechanism deposit business condition time sequence data to obtain target historical sequence data, wherein the target historical sequence data comprise sequence data with stable fluctuation. And the calculating unit is used for calculating the target test statistic of the target historical sequence data by using a unit root test algorithm. And the classification unit is used for classifying the time sequence data of the deposit service condition of the historical institution according to the target test statistic to obtain historical stationary sequence data and historical residual sequence data.
According to an embodiment of the present disclosure, the third training sub-module includes a first prediction unit, a second determination unit, and an adjustment unit. And the first prediction unit is used for inputting the first historical institution deposit business condition prediction data and the second historical institution deposit business condition prediction data into a third initial sub-model and outputting third historical institution deposit business condition prediction data. And the second determining unit is used for determining the error of the prediction result according to the third history institution deposit business condition prediction data and institution actual deposit business condition data. And the adjusting unit is used for adjusting the model parameters of the third initial submodel according to the prediction result error to obtain a result correction submodel.
According to an embodiment of the present disclosure, the adjustment unit includes a first determination subunit and a second determination subunit. And the first determining subunit is used for determining the model parameters of the third initial submodel as the target model parameters under the condition that the error of the prediction result is smaller than a preset threshold value. And the second determining subunit is used for determining a result correction submodel according to the target model parameters.
According to the embodiment of the present disclosure, any plurality of the obtaining module 510, the processing module 520, the first prediction module 530, the second prediction module 540, and the modification module 550 may be combined into one module to be implemented, or any one of the modules may be split into a plurality of modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the obtaining module 510, the processing module 520, the first predicting module 530, the second predicting module 540, and the modifying module 550 may be at least partially implemented as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented by hardware or firmware in any other reasonable manner of integrating or packaging a circuit, or implemented by any one of three implementations of software, hardware, and firmware, or any suitable combination of any of them. Alternatively, at least one of the obtaining module 510, the processing module 520, the first prediction module 530, the second prediction module 540 and the modification module 550 may be at least partially implemented as a computer program module, which when executed, may perform a corresponding function.
FIG. 6 schematically illustrates a block diagram of an electronic device suitable for implementing a method for predicting institution credit business conditions in accordance with an embodiment of the present disclosure.
As shown in fig. 6, an electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. Processor 601 may include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or associated chipset, and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), among others. The processor 601 may also include on-board memory for caching purposes. The processor 601 may comprise a single processing sub-module or a plurality of processing sub-modules for performing the different actions of the method flows according to embodiments of the present disclosure.
In the RAM 603, various programs and data necessary for the operation of the electronic apparatus 600 are stored. The processor 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. The processor 601 performs various operations of the method flows according to embodiments of the present disclosure by executing programs in the ROM 602 and/or RAM 603. Note that the programs may also be stored in one or more memories other than the ROM 602 and RAM 603. The processor 601 may also perform various operations of the method flows according to embodiments of the present disclosure by executing programs stored in the one or more memories.
The present disclosure also provides a computer-readable storage medium, which may be embodied in the device/apparatus/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.
According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, a computer-readable storage medium may include the ROM 602 and/or RAM 603 described above and/or one or more memories other than the ROM 602 and RAM 603.
Embodiments of the present disclosure also include a computer program product comprising a computer program containing program code for performing the method illustrated by the flow chart. When the computer program product runs in a computer system, the program code is used for causing the computer system to realize the method provided by the embodiment of the disclosure.
The computer program performs the above-described functions defined in the system/apparatus of the embodiments of the present disclosure when executed by the processor 601. The above-described systems, devices, modules, sub-modules, etc. may be implemented by computer program modules according to embodiments of the present disclosure.
In one embodiment, the computer program may be hosted on a tangible storage medium such as an optical storage device, a magnetic storage device, and the like. In another embodiment, the computer program may also be transmitted, distributed in the form of signals over a network medium, downloaded and installed via the communication section 609, and/or installed from a removable medium 611. The computer program containing program code may be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.
In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 609, and/or installed from the removable medium 611. The computer program, when executed by the processor 601, performs the above-described functions defined in the system of the embodiments of the present disclosure. In accordance with embodiments of the present disclosure, the systems, devices, apparatuses, modules, sub-modules, etc. described above may be implemented by computer program modules.
In accordance with embodiments of the present disclosure, program code for executing computer programs provided by embodiments of the present disclosure may be written in any combination of one or more programming languages, and in particular, these computer programs may be implemented using high level procedural and/or object oriented programming languages, and/or assembly/machine languages. The programming language includes, but is not limited to, programming languages such as Java, C + +, python, the "C" language, or the like. The program code may execute entirely on the user computing device, partly on the user device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
It will be appreciated by a person skilled in the art that various combinations or/and combinations of features recited in the various embodiments of the disclosure and/or in the claims may be made, even if such combinations or combinations are not explicitly recited in the disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present disclosure may be made without departing from the spirit or teaching of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.
The embodiments of the present disclosure have been described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described separately above, this does not mean that the measures in the embodiments cannot be used in advantageous combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be devised by those skilled in the art without departing from the scope of the disclosure, and these alternatives and modifications are intended to fall within the scope of the disclosure.
Claims (10)
1. A method for predicting an institution deposit transaction condition, comprising:
acquiring mechanism deposit business condition time sequence data in a preset time period;
carrying out differential processing on the mechanism deposit business condition time sequence data to obtain stable sequence data and residual sequence data;
inputting the stable sequence data into a first prediction sub-model in a mechanism deposit business condition prediction model, and outputting a first prediction result, wherein the first prediction result represents first mechanism deposit business condition prediction information of a target time period;
inputting the residual sequence data into a second predictor model in the institution deposit business condition prediction model, and outputting a second prediction result, wherein the second prediction result represents second institution deposit business condition prediction information of the target time period, and the model structure of the second predictor model is different from that of the first predictor model; and
and inputting the first prediction result and the second prediction result into a result correction sub-model in the institution deposit business condition prediction model, and outputting third institution deposit business condition prediction information of the target time period.
2. The method of claim 1, wherein the differentiating the institution deposit service condition time series data to obtain stationary sequence data and residual sequence data comprises:
performing differential processing on the mechanism deposit business condition time sequence data to obtain target sequence data, wherein the target sequence data comprises sequence data with stable fluctuation;
calculating a target test statistic of the target sequence data by using a unit root test algorithm;
and classifying the mechanism deposit business condition time sequence data according to the target test statistic to obtain the stable sequence data and the residual sequence data.
3. The method of claim 2, wherein the differentiating the institution deposit transaction status time-series data to obtain target sequence data comprises:
carrying out differential processing on the mechanism deposit business state time sequence data to obtain a target time sequence diagram;
and determining the time sequence data which presents stable fluctuation in the target time sequence chart as the target sequence data.
4. The method of claim 1, wherein the method of training the institution deposit business condition prediction model comprises:
acquiring a sample data set, wherein the sample data set comprises sample data of a plurality of historical preset time periods, and the sample data comprises historical institution deposit service condition time sequence data and institution actual deposit service condition data of a historical target time period;
for each sample data, carrying out differential processing on the historical deposit business condition time sequence data to obtain historical stable sequence data and historical residual error sequence data;
training a first initial sub-model according to the historical steady sequence data and the actual deposit business condition data of the organization to obtain the trained first prediction sub-model;
training a second initial sub-model according to the historical residual sequence data and the actual deposit business condition data of the organization to obtain a trained second prediction sub-model;
inputting the historical stationary sequence data into the first prediction sub-model, and outputting first historical institution deposit business condition prediction data;
inputting the historical residual sequence data into the second prediction sub-model, and outputting second historical institution deposit service condition prediction data;
and training a third initial sub-model according to the first historical institution deposit business condition prediction data, the second historical institution deposit business condition prediction data and the institution actual deposit business condition data to obtain the trained result correction sub-model.
5. The method of claim 4, wherein the training a third initial sub-model according to the first historical institutional deposit business condition prediction data, the second historical institutional deposit business condition prediction data, and the institutional actual deposit business condition data to obtain the trained result correction sub-model comprises:
inputting the predicted data of the deposit business condition of the first historical institution and the predicted data of the deposit business condition of the second historical institution into the third initial submodel, and outputting the predicted data of the deposit business condition of the third historical institution;
determining a prediction result error according to the third history institution deposit business condition prediction data and the institution actual deposit business condition data;
and adjusting the model parameters of the third initial submodel according to the predicted result error to obtain the result correction submodel.
6. The method of claim 5, wherein said adjusting model parameters of said third initial submodel according to said predicted outcome error to obtain said outcome corrected submodel comprises:
determining the model parameters of the third initial sub-model as target model parameters under the condition that the error of the prediction result is smaller than a preset threshold value;
and determining the result correction submodel according to the target model parameters.
7. The method of claim 4, wherein the differentiating the historical deposit service condition time series data to obtain historical stationary sequence data and historical residual sequence data comprises:
performing differential processing on the historical institution deposit service condition time sequence data to obtain target historical sequence data, wherein the target historical sequence data comprises sequence data with stable fluctuation;
calculating a target test statistic of the target historical sequence data by using a unit root test algorithm;
and classifying the time sequence data of the deposit service condition of the historical institution according to the target test statistic to obtain the historical stable sequence data and the historical residual sequence data.
8. An apparatus for predicting institution credit business conditions, comprising:
the acquisition module is used for acquiring the mechanism deposit business condition time sequence data in a preset time period;
the processing module is used for carrying out differential processing on the mechanism deposit business state time sequence data to obtain stable sequence data and residual sequence data;
the first prediction module is used for inputting the stable sequence data into a first prediction sub-model in the mechanism deposit business condition prediction model and outputting a first prediction result, wherein the first prediction result represents the first mechanism deposit business condition prediction information of a target time period;
the second prediction module is used for inputting the residual sequence data into a second prediction submodel in the institution deposit business condition prediction model and outputting a second prediction result, wherein the second prediction result represents second institution deposit business condition prediction information of the target time period, and the model structure of the second prediction submodel is different from that of the first prediction submodel; and
and the correction module is used for inputting the first prediction result and the second prediction result into a result correction sub-model in the institution deposit business condition prediction model and outputting third institution deposit business condition prediction information of the target time period.
9. An electronic device, comprising:
one or more processors;
a storage device for storing one or more programs,
wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-7.
10. A computer readable storage medium having stored thereon executable instructions which, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210495857.XA CN114781745A (en) | 2022-05-07 | 2022-05-07 | Method, device and equipment for predicting mechanism deposit business condition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210495857.XA CN114781745A (en) | 2022-05-07 | 2022-05-07 | Method, device and equipment for predicting mechanism deposit business condition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114781745A true CN114781745A (en) | 2022-07-22 |
Family
ID=82437172
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210495857.XA Pending CN114781745A (en) | 2022-05-07 | 2022-05-07 | Method, device and equipment for predicting mechanism deposit business condition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114781745A (en) |
-
2022
- 2022-05-07 CN CN202210495857.XA patent/CN114781745A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11037236B1 (en) | Algorithm and models for creditworthiness based on user entered data within financial management application | |
US10489865B1 (en) | Framework for cash-flow forecasting | |
CN110400184B (en) | Method and apparatus for generating information | |
CN114462532A (en) | Model training method, device, equipment and medium for predicting transaction risk | |
CN112116245A (en) | Credit risk assessment method, credit risk assessment device, computer equipment and storage medium | |
US20150081398A1 (en) | Determining a performance target setting | |
CN111401940A (en) | Feature prediction method, feature prediction device, electronic device, and storage medium | |
CN112037007A (en) | Credit approval method for small and micro enterprises and electronic equipment | |
US20220083897A1 (en) | Explanation for time series forecasting models | |
CN110766184A (en) | Order quantity prediction method and device | |
CN117934154A (en) | Transaction risk prediction method, model training method, device, equipment, medium and program product | |
CN114238993A (en) | Risk detection method, apparatus, device and medium | |
US20150088727A1 (en) | Method for determining creditworthiness for exchange of a projected, future asset | |
US20220067460A1 (en) | Variance Characterization Based on Feature Contribution | |
US20220164405A1 (en) | Intelligent machine learning content selection platform | |
CN115408297A (en) | Test method, device, equipment and medium | |
CN113129127A (en) | Early warning method and device | |
CN114781745A (en) | Method, device and equipment for predicting mechanism deposit business condition | |
US20210248512A1 (en) | Intelligent machine learning recommendation platform | |
CN115048561A (en) | Recommendation information determination method and device, electronic equipment and readable storage medium | |
CN114708081A (en) | Credit risk prediction method and device, electronic equipment and readable storage medium | |
CN114816955A (en) | Database performance prediction method and device | |
CN112132689A (en) | Recommendation method and device based on time sequence factor event | |
CN113052509A (en) | Model evaluation method, model evaluation apparatus, electronic device, and storage medium | |
US20240242269A1 (en) | Utilizing a deposit transaction predictor model to determine future network transactions |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |